RiemStiefel: Tools for Inference, Learning, and Optimization on Stiefel...

In RiemStiefel: Inference, Learning, and Optimization on Stiefel Manifold

Description

Stiefel manifold St(p,r) is the set of all orthonormal r-frames in R^p, which is indeed a Riemannian manifold. For X \in St(p,r), it is characterized as

X^\top X = I_{r \times r}

. We provide algorithms for statistical inference, optimization, and learning over the Stiefel manifold. In our package, we use a convention to represent each data point on Stiefel manifold St(p,r) as (p\times r) matrix.

RiemStiefel documentation built on March 26, 2020, 7:48 p.m.